Extrapolation Puts Big Data to Work for the Little Guy

Large frauds can generate hundreds of thousands of false claims for payment. The whistleblowers who uncover such frauds rarely have proof of every single false claim. Instead, they will usually have evidence of the overarching scheme—from emails, company memos, or conversations with insiders—along with some examples of the false claims that the scheme has generated. Now more than ever, that could be enough to win, as long as the whistleblower and their lawyers know how to use inferential statistics to their advantage.

Suppose that two electrical engineers find out that their employer, a defense contractor, used Metal D for electrical components in 10,000 missile-guidance systems, when the contract called for Metal A. They know that Metal D degrades ten times faster than Metal A, and they have company emails showing that their employer intentionally switched the metals to boost profits. It would be impossible for them to gather proof for each of the 10,000 guidance systems, but they may not have to. Federal courts are increasingly willing to accept proof of the overall scheme, combined with a statistical sample of the false claims that the scheme has generated.

The latest decision approving sampling and extrapolation is United States ex rel. Integra Med Analytics, LLC v. Creative Solutions in Healthcare, Inc., No. 17-cv-1249, (W.D. Tex. Nov. 13, 2019). The defendant in that case is a skilled nursing facility. The plaintiff relied on witness interviews and statistical analyses to allege that the defendant had fraudulently submitted more than $96 million in false claims to Medicare and Medicaid. The defendant tried to dismiss the case on the grounds that the plaintiff was relying on statistics rather than claim-by-claim proof.  But the district court denied its motion, holding that “the statistics in this case form the ‘reliable indicia’ that, when paired with the witness interviews, lead to a ‘strong inference’ that Defendant submitted false claims.”

The lesson is this: when whistleblowers uncover large, complex frauds that have generated many separate false claims, they should immediately consult with experienced counsel to see if statistical proof can help them win their case.

Comment